GAS, COAL 563 



h is a ground glass globo enclosing the whole arrangement, and having only an 

 opening below for the admission of air to the flame. In order to dispense with tho 

 descending tube, to which there are some objections, Mr. Rutter has constructed a 

 ventilating burner in which the ordinary glass chimney is made to terminate in a 

 metal tube, through which the products of combustion are conveyed away. Mr. Dixon 

 has also constructed a modification of Faraday's burner, the peculiarity of which con- 

 sists in the use of a separate tube bringing air to the flame from the same place, 

 outside tho building, to which the products of the burner are conveyed ; this con- 

 trivance is said to prevent downward draughts through the escape-pipe, and a conse- 

 quently unsteady flame. 



ON THE ESTIMATION OF THE VALUE OF ILLUMINATING GAS. 



There are two methods in uso for estimating the illuminating value of gas, viz. 



1st. Tho photometric method. 



2nd. Chemical analysis. 



The photometric method consists in comparing the intensity of the light emitted 

 by a gas flame, consuming a known volume of gas, with that yielded by some other 

 source of light taken as a standard. The standard employed is usually a spermaceti 

 candle, burning at the rate of 120 grains of sperm per hour. A spermaceti can-lie of 

 six to the pound usually burns a*a somewhat quicker rate than this ; but in all cases 

 the consumption of sperm by the candle during the course of each experiment ought 

 to be carefully ascertained by weighing, and the results obtained corrected to the 120- 

 grain standard. Thus, suppose that during an experiment the consumption of sperm 

 was at the rate of 130 grains per hour, and that the gas flame being tested gave a 

 light equal to 20 such candles, and. it is required to know the light of this flame in 

 standard 120 grain candles, then 



120 : 130 :: 20 :: 217, 



or, 20 candles burning at the rate of 130 grains per hour, are equal to 217 candles 

 burning at the rate of 120 grains per hour. 



There are two methods of estimating the comparative intensity of the light of the 

 gas and candle flames, both founded upon the optical law that the intensity of light 

 diminishes in the inverse ratio of the square of the distance from its source. Thus, 

 if a sheet of writing paper be held at the distance of one foot from a candle, so that 

 its surface is perpendicular to a line joining the centre of the sheet and the flame, it 

 will be illuminated with a light four times as intense as that which would fall upon a 

 sheet of paper held in the same position at a distance of 2 feet ; whilst at a distance 

 of 3 feet the light would have but |th of the intensity which it possessed at 1 foot. One 

 method of estimating the comparative intensity of the gas and candle flames, consists 

 in placing the two lights and an opaque rod nearly in a straight line, and in such a 

 way as to cause each light to project a shadow of the rod upon a white screen placed 

 at a distance of about 1 foot behind the rod. The two shadows must now be 

 rendered of equal intensity by moving the candle either nearer to the rod or further 

 from it. The shadows will be of equal intensity when the light falling upon the 

 white screen from both sources is equal ; and if now the respective distances of the 

 candle and gas flame from the screen be measured, then the square of the distance of 

 the gas flame divided by the square of the distance of the candle will give the 

 illuminating power of the gas in candles. Thus, if equally intense shadows fall upon 

 the screen when the candle is 3 feet distant and the gas flame 12 feet, the illuminating 

 power of the gas flame will be 



^?-i- 16 candles. 



This method of estimating the illuminating power of a gas flame, known as the 

 shadow test, is very easy of execution, and would appear from the description to be 

 capable of yielding results of considerable accuracy; nevertheless, an unexpected 

 difficulty arises from the great difference in colour of the two shadows ; that of the 

 gas being of a bluish brown, whilst that of the candle is of a yellow-brown tinge. 

 This difference of tint renders it exceedingly difficult for the observer to ascertain 

 when the two shadows possess equal intensity ; and, consequently, tho limits of error 

 attending determinations by this test are probably, even in the hands of an experienced 

 operator, never less than 5 per cent., and frequently even as much as 10 per cent. 

 The shadow test has, therefore, been all but superseded by the Bunsen's Photometer, 

 which consists of a graduated metal or wooden rod about 8 or 10 feet long, and 

 sufficiently strong to be inflexible. At one extremity of this rod is placed the gas 



oo2 



